Debugging using GDB and gdbserver

Debugging using GDB and gdbserver

[Justin Wojdacki]

How does GDB work under MIPS Linux? I'm trying to do a bring-up of an
embedded device, and it looks like the kernel is missing the code
needed to handle software breakpoints. Are there patches that need to
be applied to the kernel? 

-- 
-------------------------------------------------
Justin Wojdacki        
justin.wojdacki@analog.com         (408) 350-5032
Communications Processors Group -- Analog Devices

Re: Debugging using GDB and gdbserver

[Daniel Jacobowitz]

On Sat, Jun 15, 2002 at 01:01:24PM -0700, Justin Wojdacki wrote:
> 
> How does GDB work under MIPS Linux? I'm trying to do a bring-up of an
> embedded device, and it looks like the kernel is missing the code
> needed to handle software breakpoints. Are there patches that need to
> be applied to the kernel? 

No.  If you use a current GDB (I recommend 5.2 or CVS) it should work
just fine, if you are using a recent kernel (you didn't mention what
version you were looking at).

-- 
Daniel Jacobowitz                           Debian GNU/Linux Developer
MontaVista Software                         Carnegie Mellon University

Re: Debugging using GDB and gdbserver

[Justin Wojdacki]

Daniel Jacobowitz wrote:
> 
> On Sat, Jun 15, 2002 at 01:01:24PM -0700, Justin Wojdacki wrote:
> >
> > How does GDB work under MIPS Linux? I'm trying to do a bring-up of an
> > embedded device, and it looks like the kernel is missing the code
> > needed to handle software breakpoints. Are there patches that need to
> > be applied to the kernel?
> 
> No.  If you use a current GDB (I recommend 5.2 or CVS) it should work
> just fine, if you are using a recent kernel (you didn't mention what
> version you were looking at).
> 
> --
> Daniel Jacobowitz                           Debian GNU/Linux Developer
> MontaVista Software                         Carnegie Mellon University

Sorry, I'm using the 2.4.10 kernel and GDB 5.2. What I see happening
is the BREAK 5 instruction from a software breakpoint is hit, and the
kernel loops continuously on that, as it appears to have no way to
deal with that exception. I'm running gdbserver on the MIPS target and
gdb as a cross-debugger on an x86 host (RedHat 7.1). To me, it looks
like when the debugging breakpoint is hit, gdbserver should get
scheduled to run and handle the breakpoint, but instead the child
keep's getting scheduled. 

-- 
-------------------------------------------------
Justin Wojdacki        
justin.wojdacki@analog.com         (408) 350-5032
Communications Processors Group -- Analog Devices

Re: Debugging using GDB and gdbserver

[Daniel Jacobowitz]

On Sat, Jun 15, 2002 at 01:29:56PM -0700, Justin Wojdacki wrote:
> Daniel Jacobowitz wrote:
> > 
> > On Sat, Jun 15, 2002 at 01:01:24PM -0700, Justin Wojdacki wrote:
> > >
> > > How does GDB work under MIPS Linux? I'm trying to do a bring-up of an
> > > embedded device, and it looks like the kernel is missing the code
> > > needed to handle software breakpoints. Are there patches that need to
> > > be applied to the kernel?
> > 
> > No.  If you use a current GDB (I recommend 5.2 or CVS) it should work
> > just fine, if you are using a recent kernel (you didn't mention what
> > version you were looking at).
> > 
> > --
> > Daniel Jacobowitz                           Debian GNU/Linux Developer
> > MontaVista Software                         Carnegie Mellon University
> 
> Sorry, I'm using the 2.4.10 kernel and GDB 5.2. What I see happening
> is the BREAK 5 instruction from a software breakpoint is hit, and the
> kernel loops continuously on that, as it appears to have no way to
> deal with that exception. I'm running gdbserver on the MIPS target and
> gdb as a cross-debugger on an x86 host (RedHat 7.1). To me, it looks
> like when the debugging breakpoint is hit, gdbserver should get
> scheduled to run and handle the breakpoint, but instead the child
> keep's getting scheduled. 

Software breakpoints have worked at least as far back as 2.4.2.  This
most likely means that the exception handling for your board is broken.

-- 
Daniel Jacobowitz                           Debian GNU/Linux Developer
MontaVista Software                         Carnegie Mellon University

Re: Debugging using GDB and gdbserver

[Justin Wojdacki]

Daniel Jacobowitz wrote:
> 
> Software breakpoints have worked at least as far back as 2.4.2.  This
> most likely means that the exception handling for your board is broken.
> 
> --
> Daniel Jacobowitz                           Debian GNU/Linux Developer
> MontaVista Software                         Carnegie Mellon University

What exception handling is the board expected to provide? 

-- 
-------------------------------------------------
Justin Wojdacki        
justin.wojdacki@analog.com         (408) 350-5032
Communications Processors Group -- Analog Devices

Re: Debugging using GDB and gdbserver

[Justin Wojdacki]

Daniel Jacobowitz wrote:
> 
> Software breakpoints have worked at least as far back as 2.4.2.  This
> most likely means that the exception handling for your board is broken.
> 

Sorry, originally misinterpretted your use of "board" as referring to
the board itself, and perhaps PMON (I've found a number of references
online to GDB talking to PMON, but not much else). 

So what I've found by looking at other board-specific code revolves
around GDB talking to an in-kernel stub via the serial port. As the
board I'm working with has an unreliable serial port (and some
incarnations don't even have that), what about ethernet-based
debugging? Is that do-able, say via putDebugChar() (although I suspect
this poses an initialization problem)? 

Thanks for the info so far. 

-- 
-------------------------------------------------
Justin Wojdacki        
justin.wojdacki@analog.com         (408) 350-5032
Communications Processors Group -- Analog Devices

Re: Debugging using GDB and gdbserver

[Daniel Jacobowitz]

On Sat, Jun 15, 2002 at 03:16:05PM -0700, Justin Wojdacki wrote:
> Daniel Jacobowitz wrote:
> > 
> > Software breakpoints have worked at least as far back as 2.4.2.  This
> > most likely means that the exception handling for your board is broken.
> > 
> 
> Sorry, originally misinterpretted your use of "board" as referring to
> the board itself, and perhaps PMON (I've found a number of references
> online to GDB talking to PMON, but not much else). 
> 
> So what I've found by looking at other board-specific code revolves
> around GDB talking to an in-kernel stub via the serial port. As the
> board I'm working with has an unreliable serial port (and some
> incarnations don't even have that), what about ethernet-based
> debugging? Is that do-able, say via putDebugChar() (although I suspect
> this poses an initialization problem)? 
> 
> Thanks for the info so far. 

Wait, wait.  What are you trying to do?  Originally you were talking
about userspace debugging via gdbserver.  Now you're talking about
kernel debugging via kgdb.  They're separate (and coexisting can cause
problems if you are not careful with your exception handlers; I do not
remember when my patches to make that work went into the tree, or if
someone else did it).

gdbserver can just use TCP.

-- 
Daniel Jacobowitz                           Debian GNU/Linux Developer
MontaVista Software                         Carnegie Mellon University

Re: Debugging using GDB and gdbserver

[Justin Wojdacki]

Daniel Jacobowitz wrote:
>
> Wait, wait.  What are you trying to do?  Originally you were talking
> about userspace debugging via gdbserver.  Now you're talking about
> kernel debugging via kgdb.  They're separate (and coexisting can cause
> problems if you are not careful with your exception handlers; I do not
> remember when my patches to make that work went into the tree, or if
> someone else did it).
> 
> gdbserver can just use TCP.
> 
> --
> Daniel Jacobowitz                           Debian GNU/Linux Developer
> MontaVista Software                         Carnegie Mellon University

Sorry for the confusion. I've been discussing userspace debugging via
gdbserver the entire time. However, I've noticed that gdbserver
doesn't seem to be fully functional because the kernel doesn't seem to
be handling the "BREAK 5" instruction correctly. You mentioned
problems with board exception handling and I looked at the ddb series
board support code. In there, I found handling for software
breakpoints, and got the impression from the code that it was a
general debugging interface, not just for kernel debugging. Again,
sorry for the confusion.

As it stands right now, when I get hit a "BREAK 5" instruction,
gdbserver never get's a chance to handle it, as the kernel keeps
scheduling the child process I'm trying to debug, and hitting the
"BREAK 5" instruction over and over again. What I can't seem to find
out is how gdbserver is supposed to get scheduled again. 

-- 
-------------------------------------------------
Justin Wojdacki        
justin.wojdacki@analog.com         (408) 350-5032
Communications Processors Group -- Analog Devices

Re: Debugging using GDB and gdbserver

[Daniel Jacobowitz]

On Sat, Jun 15, 2002 at 03:40:08PM -0700, Justin Wojdacki wrote:
> Daniel Jacobowitz wrote:
> >
> > Wait, wait.  What are you trying to do?  Originally you were talking
> > about userspace debugging via gdbserver.  Now you're talking about
> > kernel debugging via kgdb.  They're separate (and coexisting can cause
> > problems if you are not careful with your exception handlers; I do not
> > remember when my patches to make that work went into the tree, or if
> > someone else did it).
> > 
> > gdbserver can just use TCP.
> > 
> > --
> > Daniel Jacobowitz                           Debian GNU/Linux Developer
> > MontaVista Software                         Carnegie Mellon University
> 
> Sorry for the confusion. I've been discussing userspace debugging via
> gdbserver the entire time. However, I've noticed that gdbserver
> doesn't seem to be fully functional because the kernel doesn't seem to
> be handling the "BREAK 5" instruction correctly. You mentioned
> problems with board exception handling and I looked at the ddb series
> board support code. In there, I found handling for software
> breakpoints, and got the impression from the code that it was a
> general debugging interface, not just for kernel debugging. Again,
> sorry for the confusion.
> 
> As it stands right now, when I get hit a "BREAK 5" instruction,
> gdbserver never get's a chance to handle it, as the kernel keeps
> scheduling the child process I'm trying to debug, and hitting the
> "BREAK 5" instruction over and over again. What I can't seem to find
> out is how gdbserver is supposed to get scheduled again. 

What should happen is that the child receives a signal (SIGTRAP) after
the exception.  Then it is scheduled again, drops into do_signal, and
the kernel notices that the traced bit is set and wakes the tracer.  I'd
guess your board needs to do something different to deliver the SIGTRAP
properly, if that isn't happening.

-- 
Daniel Jacobowitz                           Debian GNU/Linux Developer
MontaVista Software                         Carnegie Mellon University

Re: Debugging using GDB and gdbserver

[Justin Wojdacki]

Daniel Jacobowitz wrote:
> 
> What should happen is that the child receives a signal (SIGTRAP) after
> the exception.  Then it is scheduled again, drops into do_signal, and
> the kernel notices that the traced bit is set and wakes the tracer.  I'd
> guess your board needs to do something different to deliver the SIGTRAP
> properly, if that isn't happening.
> 
> --
> Daniel Jacobowitz                           Debian GNU/Linux Developer
> MontaVista Software                         Carnegie Mellon University

Okay, thanks for the clarification. :)

-- 
-------------------------------------------------
Justin Wojdacki        
justin.wojdacki@analog.com         (408) 350-5032
Communications Processors Group -- Analog Devices